Sidekiq Version Check fails

support
1

Hi,

I´m running discourse via docker - version 2.0.0.beta1 +4. A few days ago the version:check of sidekiq failed:

Job exception: getaddrinfo: Temporary failure in name resolution (SocketError)
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/socket.rb:101:in `getaddrinfo'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/socket.rb:101:in `connect'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/ssl_socket.rb:154:in `connect'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/socket.rb:29:in `initialize'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/ssl_socket.rb:9:in `initialize'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/connection.rb:403:in `new'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/connection.rb:403:in `socket'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/connection.rb:100:in `request_call'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/middlewares/mock.rb:48:in `request_call'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/middlewares/instrumentor.rb:26:in `request_call'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/middlewares/base.rb:16:in `request_call'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/middlewares/base.rb:16:in `request_call'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/middlewares/base.rb:16:in `request_call'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon/connection.rb:249:in `request'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/excon-0.56.0/lib/excon.rb:239:in `get'
/var/www/discourse/lib/discourse_hub.rb:42:in `singular_action'
/var/www/discourse/lib/discourse_hub.rb:26:in `get'
/var/www/discourse/lib/discourse_hub.rb:14:in `discourse_version_check'
/var/www/discourse/app/jobs/scheduled/version_check.rb:13:in `execute'
/var/www/discourse/app/jobs/base.rb:134:in `block (2 levels) in perform'
/var/www/discourse/vendor/bundle/ruby/2.4.0/gems/rails_multisite-1.1.2/lib/rails_multisite/connection_management.rb:77:in `with_connection'
/var/www/discourse/app/jobs/base.rb:129:in `block in perform'
/var/www/discourse/app/jobs/base.rb:125:in `each'
/var/www/discourse/app/jobs/base.rb:125:in `perform'
/var/www/discourse/app/jobs/base.rb:180:in `perform'
/var/www/discourse/lib/scheduler/manager.rb:91:in `process_queue'
/var/www/discourse/lib/scheduler/manager.rb:37:in `block in initialize'

Can anyone tell me please how to further debug it or solve this problem? It could be related to a debian upgrade from 8 to 9.

Regards,

ben

2

That’s a DNS failure. It may just be transient. Test if name resolution works from within and without containers on the server.

3 Likes
3

Thanks, can you please tell me example commands to test it within and without the container?

4

The command curl -I https://api.discourse.org/api/version_check should return something like:

HTTP/1.1 200 OK
Server: nginx
Date: Fri, 26 Jan 2018 15:58:15 GMT
Content-Type: application/json; charset=utf-8
X-Frame-Options: SAMEORIGIN
X-XSS-Protection: 1; mode=block
X-Content-Type-Options: nosniff
ETag: W/"2225ccc5f2df97fbda0cb43244c63a97"
Cache-Control: max-age=0, private, must-revalidate
X-Request-Id: d3357a4f-dfb8-4290-8811-1b71d9cd7284
X-Runtime: 0.006124
X-UA-Compatible: IE=edge
Discourse-Proxy-ID: corp-router-tiehunter02
1 Like